Contribution No. 34

SOUR HYDROCARBONS - THE ELASTOMER CHALLENGE'

D.L. Hertz, Jr. Seals Eastern Inc.

Red Bank, N.J. 07701

Presented at a meeting of the Rubber Division, American Chemical Society Las Vegas, Nevada, May 20-23, 1980

INTRODUCTION

Hydrofluorocarbon elastomers, ASTM designation D1418-FKM, are

produced domestically by Dupont and 3M Company under the trade names of

Viton and Fluorel. FKM elastomers have been extensively studied, and ample

published information is available both from suppliers and from journals

such as Rubber Chemistry and Technology (RC&T). This paper deals primarily

with interactions resulting from increasing weight-per cent of fluorine and

increasing molecular weight of FKM elastomers in the "sour" hydrocarbon

environment, not previously reported. With a few exceptions, most of the

data utilized are available from RC&T and the Rubber Division's book

"Science and Technology of Rubber."

EXPERIMENTAL

Early work on FKM formulations had indicated high molecular weight polymers

(FKM1C) had superior mechanical properties in gas-liquid refrigeration

applications. Initial testing in the sour crude environment confirmed this

observation. We believed increasing weight-per cent fluorine elastomers would show

further improvements, a fallacious assumption.

Backbone cleavage was presumed and subsequently proven as another

incorrect assumption1. This led to a more rigorous evaluation of molecular

weight relationships and physical modulus characteristics, key properties of

the original FKM seal compounds.

DISCUSSION AND RESULTS

Elastomers-The Panorama

Mechanical properties.-"The strength and extensibility of an elastomer

depend on its overall viscoelastic properties." This is part of Smith's

introductory statement in his paper, "Strength of Elastomers-A Perspective."2

Table I from this work neatly defines the elastomer we are studying. Strength

and toughness are desired physical characteristics, one created by molecular

mobility and the other by molecular immobility, a two phase condition, not

available in hydrofluorocarbon (FKM) elastomers. FKM elastomers are categorized

according to Smith as: filled, single-phase, non-crystalline materials.

Physical state and temperature.-Figure 13 is an idealized plot of shear

modulus versus temperature. This curve represents all single phase,

non-crystalline, filled or unfilled, crosslinked elastomers. The highest

shear value represents the fracture energy of a typical crystalline material.

The 1000 fold decrease in shear modulus with increasing temperature is a

typical characteristic of this category of elastomers. Logically, increasing

molecular weight increases the melt temperature, and crosslinking extends the

rubbery plateau.

Figure 2, modulus (Young's modulus E) vs. temperature is a plot of real

world FKM1 data4 representing the circled area in Figure 1. Shear modulus is 1/3

to 1/4 Young's modulus (dependent on filler loading). This transition zone should

be avoided in dynamic applications.

The dramatic degree of change in shear modulus versus temperature

points out the viscoelastic nature of elastomers. A more comprehensive

background is furnished by reviewing the failure envelope concept:

Failure envelope concept.-Smith5,6, in some landmark work, developed his

concept of ultimate properties, a major contribution. Earlier, William's, Landel,

and Ferry, with their WLF equation, pointed out that all polymers, regardless of

their chemical structure, will exhibit similar viscoelastic behavior at equal

temperature intervals (T-Tg) above their respective glass-transition temperatures.

Smith has added to this concept by showing that the ultimate tensile properties of

a non-reinforced, amorphous, crosslinked elastomer can be characterized by a

failure envelope which is independent of time (strain rate) and temperature.

Figure 3 from Reference 5 schematically illustrates the general effect of

strain rate and temperature on the tensile, stress-strain properties of amorphous

elastomers. The lines originating from 0 represent stress-strain curves determined

at various strain rates and temperatures. The envelope ABC connects the rupture

points. The rupture point moves counter clockwise around the "failure envelope" as

either the strain rate is increased or the test temperature is decreased.

OA represents classical stress-strain behavior. DE and DF represent

stress-relaxation and creep terminating in an equilibrium state. The dotted lines

from G represent stress relaxation and creep terminating in a potential rupture

mode.

The stress-strain curves represent the nonlinear viscoelastic response of

an amorphous elastomer to an imposed strain, increasing directly proportionally to

time.

In more recent work, Landel7 shows an interesting series of FKM

elastomer stress-strain curves (Figure 4) plotted logarithmically. The curves,

progressively displaced upward as the temperature is decreased,

3

are terminated at a point representing rupture. The family of curves

represents a decreasing temperature run at one strain rate with the

temperature normalized to Kelvin. By plotting an envelope around the rupture

points, the curve in (Figure 3) is now emerging.

Failure envelope.-Smith's next approach was to run a non-reinforced FKM2

elastomer and plot the log of stress and strain at break, as shown in (Figure 5)

(as opposed to the previous (Figure 4), which was plotted as the log of stress

versus strain). The data points reflect 10 strain rates (from 0.02 to 20 inches per

minute) for each of the nine temperatures noted. Again, the familiar parabolic

curve is apparent. The data scatter in the low temperature area is understandable

in light of the Tg curve (Figure 1) and the Meyer-Ferri curve (Figure 10). Both

figures indicate that the rubbery characteristic is becoming "blurred" in this

temperature range because microcrystallinity is appearing.

Filler reinforcement.-Table I points out the contribution of fillers to the

source of strength of elastomers. At elevated temperatures, the contribution

becomes greatly minimized8 as evidenced by Figures 69 and 710. The data, developed

by Greensmith, Mullins, and Thomas11 and plotted by Gent, represents critical

fracture energy Gc, or critical strain energy release. The validity of a high

temperature design model based on stress-strain properties for an unfilled,

single-phase, non-crystallizable elastomer is more apparent after reviewing these

figures. There is an obvious contribution by specific fillers at higher

temperatures, but it is the contribution toward higher shear modulus rather than

true stress at break.

4

Crosslink density.-Increasing crosslink density affects physical properties in the

generalized manner described by Figure 812. Dependence of true stress at break for an FKM1C

(highest molecular weight) elastomer with varying diamine curative levels is illustrated by

Figure 913. The diamines are not as efficient as the newer bisphenols. The physical

contribution of chain entanglements is very apparent. The peak stress at break values are

formulations that would be considered "undercured" by conventional compounding standards.

Supplier data14 and other references15,16 give ample technical details of crosslinking

chemistry. Our interest is primarily in the development of higher-modulus while optimizing

other various critical physical properties such as tear strength and compression set.

Theory

Rubber elasticity.-"It is now well established that the stress in a

deformed rubber originates within the chains of the network. Inter-chain

interactions contribute negligibly to the stress ."17 (Figure 10) from

Reference 18 illustrates the Meyer-Ferri experiment which helped to establish

the importance of entropy as the deformation mechanism in rubber. The third

law of thermodynamics is essentially satisfied by this experiment,

illustrating that force is directly proportional to temperature using absolute

(Kelvin) temperature as a base. Thermodynamics is thus firmly established as a

contributing discipline for the understanding of other polymeric responses.

Intermolecular forces.-Intermolecular forces are the attractive

forces between two molecules due to electrical imbalances. We are

5

considering the three major intermolecular force components which

control physical strength, thermal properties, and solubility

(compatibility).

Dispersion (London) forces.-δd , are common to all matter. They are

omni-directional and are a major contributor to the physical properties of

materials. They are the only intermolecular force in non-polar materials. The

individual attractive effect is relatively insignificant but cumulative with

increasing molecular weight. Table II illustrates that a change in physical

state from a gas, to a liquid, to a solid, is the direct result of dispersion

forces.

Dipole (Keesom) forces. -δp , also called orientation-electrostatic forces,

are created when the vector sums of the various bond angles do not cancel,

creating a net dipole. Methanol is an interesting model: two gases combine to

create a liquid, (Figure 11). Dipole forces have a strong orientation effect often

creating geometric species such as dimers, trimers, tetramers, and pseudopolymers.

The model illustrated is complicated by a third intermolecular force, the hydrogen

bond, Figure 12.

Charge transfer forces. -δh , is a catch-all contribution (including hydrogen

bonds) of a series of smaller intermolecular forces, all relatively directional

like the dipole force. An interesting effect is illustrated in (Figure 13). CH4 and

CF4 are symmetrical, non-polar molecules, the boiling points dictated by dispersion

forces alone, and the difference due to different CH and CF binding energies.

Fluorine substitutions of hydrogen raises the boiling point due to H-F

interactions. The highest boiling point is for CH2F2, illustrating the highly

directional nature of the hydrogen bond.

6

Weight-Per Cent Fluorine

PTFE polymers.-PTFE (polytetrafluoroethylene) was the first totally

fluorinated (perfluoro)/polymer (76 weight-% fluorine). PTFE represents the

maximum chemical and thermal stability available in an aliphatic carbon

backbone polymer. The "folded chain" molecular architecture, Figure 1419,

precludes a rubbery phase because of the high per cent of microcrystallinity

(50-70%).

FKM elastomers.-The rubbery analogs, FKM elastomers, are created by

polymerization of two or more fluorinated monomers subsequently referred to as

structural units such as vinylidene fluoride (VF2), tetrafluoroethylene (PTFE),

hexafluoropropylene (HFP), and perfluorovinylmethylether (PVME). The bulky

side groups on the HFP (-CF3) and PVME (-OCF3) do not allow folded chain

structures, and the random coil structure results. A Tg below room

temperature, less than 30% microcrystallinity and a random coil structure are

minimum requirements for elastomers as we know them.20

Fluorine.-Weight-per cent fluorine in FKM elastomers ranges from 66%

to 73.9%. Table III shows an estimated weight-per cent of both fluorine and

hydrogen for the four generally utilized elastomers.

Hydrogen fluorine number ratio is also noted. Molecular weight

differences (19 versus 1) understates the effect of hydrogen, which has a

multiple role in FKM1-3 elastomers;

7

Hydrogen - Good effects

Cure site for dehydrohalogenation mechanism.

Improves low temperature flexibility.

Better physical properties at intermediate temperatures.

Hydrogen - Bad effects

Lowers thermal stability.

Attracts hydrogen bonding molecules (methanol, etc.).

Inherent site for acid-base interactions.

Temperature-Fluorine %.-Increasing weight-per cent fluorine raises the Tg

because of molecular size and electronegativity differences, Table IV (hydrogen is

strongly positive, fluorine strongly negative). Tg is the point of change in the

slope of the curve of volume versus temperature, Figure 1521. It is generally

accepted that all single phase, amorphous polymers have 0.025 (21/2%) free volume

at Tg22. Comparison of physical properties of most amorphous, single-phase

elastomers at equivalent temperatures corrected to Tg should give similar results.

The WLF equation can be used to predict empirically, with surprising accuracy,

property shifts up to 100°C above Tg.

Molecular Weight

Degree of polymerization.-Single-phase elastomers usually reach optimum

processing properties at about 1000 structural units, Figure 1623. Lower and

higher degrees of polymerization (molecular weight) are often available, with a

range of 500 to 2500 units not uncommon. Some typical estimated degrees of

polymerization are listed in Table V. A plot, Figure 17, of four FKM1 polymers

having increasing Mooney viscosity

8

(molecular weight) shows improvements in tensile, elongation, and compression

set. M100 modulus and durometer are relatively unchanged.

Compression set improves with increasing degree of polymerization

because of chain entanglements.

Physical entanglements.-Additional entanglements are created

by crosslinking, Figure 1825. Kramer and Ferry26 note that a typical EPM (56 mole

% ethylene) has a physical entanglement every 50 structural units. Polystyrene

has a physical entanglement every 175 structural units. FKM1 polymers might have

a physical entanglement every 150 structural units, and FFKM, every 175

structural units (due to higher bulky fluorine ratios).

The concept of physical entanglements does not lend itself to classical

solutions. High molecular weight crosslinked polymers have properties that

normally would be predicted by a crosslink density twice as high as the known

value.

Molecular weight effects.-The effect of increasing molecular weight

on tensile strength is apparent at higher temperatures. Noted later is the

effect on gas-liquid-elastomer interactions resulting in improved resistance

to blistering and fracture.

Gas-Liquid-Elastomer Interactions

Failure modes.-The failure modes of rubber components through

sponging, blistering, and rupturing are logically caused by diffusion of

gases into the elastomer. The idea of diffusion is intuitive and is readily

acceptable, in view of all elastomers having 21/2% void area at Tg, noted

earlier. Van Amerongen's classic work on diffusion27 points out that the

diffusion coefficient is dependent on molecular

9

construction, pressure, solubility and Tg. Simplistically, we will define

gas-liquid-elastomer interactions by stating: DIFFUSION creates the

problem, and SOLUBILITY exaggerates the problem.

Diffusion.-Diffusion28 is the net transport of material within a single

phase in the absence of mixing. Experiment and theory have shown that diffusion

can result from pressure gradients (pressure diffusion), temperature gradients

(thermal diffusion), external force fields (forced diffusion), and concentration

gradient diffusion.

Solubility.-Potential solubility of the C02 and H2S gases in

elastomers was reviewed by the solubility parameter concept.

Solubility parameter concept.-Data for solubility parameters, δ , was

developed on the basis of the Hildebrand29,30 regular solution theory,

using the Hansen31 modification. This modification considers δ to be the sum of three component forces: δ2 = δ2

d + δ2p + δ2

h where

δd = dispersion (London) - common to all matter,

δp = dipole (Keesom) forces - created by permanent dipoles, and

δh = charge transfer forces (including hydrogen bonds).

Hildebrand states that a difference of four or less δ units {(J/m-3)1/2.10-3} between

solvent and solute indicates a potential solubility interaction. Solubility

parameter data for Table VI and Table VII were derived by using the molecular

additive constants technique as advanced by Fedors32 from unpublished data by

Beerbower33.

C02 solubility.-C02 with δ =15.1 has a theoretical compatibility mix with FKM1

through FFKM-4, appearing more soluble with increasing weight-per cent fluorine. In

terms of molecular thermodynamics, it has a doubly degenerate bending mode34 (meaning

that it vibrates both perpendicular

10

and parallel to a plane surface), which makes it a very mobile molecule.

H2S solubility.-H2S with δ =22.6 also shows theoretical compatibility

decreasing with increasing weight-per cent fluorine. Previous testing by Seals

Eastern had indicated that increasing weight-per cent FKM elastomers equivalently

compounded were less resistant to sour crude blistering.

Conclusion

The contradictory results on H2S (theoretical solubility versus actual

data) lead us to believe we had both a physical problem of diffusion and a

thermodynamic conflict.

Physical problem-layers.-There is relative agreement that a

solvent-solute diffusion mechanism for liquids is one or more diffusion fluxes

created as gas-liquid transports into the elastomer. This phase phenomenon is

described variously as:

"Plane of no net molal flow" for binary mixtures28 ,

the thermodynamic concept of a hypothetical fixed

reference plane, Figure 20,

"Integral surface layer concept of sublayers"35 due to

solvent-solute interaction (Ueberreiter-1968),

"Membrane osmometry concept"36 whereby elastomer serves as

both membrane (surface) and solute (body).

In all cases, the equilibrium gas-liquid combination diffuses into the

solute (elastomer) until equilibrium is again attained. A change in equilibrium

causes a reverse flow, rate dependent on solubility. The boundary layer, the weak

member, is dependent on molecular weight for physical strength through chain

entanglements.

11

Failure Mechanisms

Critical flaw sites37. - Natural flaws, 40 µm or less, inevitably occur

in elastomers. Tear propagates from these sites when a large enough stress

is applied. Many compounding ingredients (Ex.-litharge-a typical FKM acid accepter)

have particle sizes in this range so potential fracture preconditions are well

established.

Table VIII38 is a comprehensive cross reference offering a perspective

of size relationships between elastomers, fillers, and other ingredients.

Assuming a natural flaw site, Figure 1939, is occupied by diffused gas, an

equilibrium change creates a positive internal pressure. At a critical pressure

(Pc), the result is a probable blister or rupture mode occurrence.

Blistering.- Blistering, usually associated with a highly elastic

solid, results when the following conditions are met:

a. gas supersaturated,

b. durometer-low (low shear modulus),

c. crosslink density-low,

d. elongation-high (>200%).

"Equation(1)" appears to adequately describe this occurrence

(E=Young's modulus).

Pc = 5E/6 (1)

Fracture.-Fracture (rupture) is the failure mode of a linear elastic

solid. Gent suggests this mechanism occurs when the following preconditions

exist:

12

a. durometer-high (high shear modulus),

b. crosslink density-high,

c. elongation-low (<100%)

"Equation(2)" is suggested by Gent to numerically define this mode

of fracture.

Pc = 4/3 λbσb (2)

Log λbσb (true stress at break) values are available from Figure 9. The

rupture failure mode is typical of high modulus seals. The surface "flaking" or

lenticular failure probably occurs with an equilibrium shift before steady state

diffusion occurs. The deeper internal fissuring, normally parallel to the

diffusion plane, is assumed to initiate at a natural flaw site near the "plane of

no net molal flow", (Figure 2040), during an equilibrium shift. The failure

proceeds as a high speed propagation mechanism41 until there is sufficient

internal energy dissipation to arrest the crack growth.

CONCLUSIONS

Diffusion of gas into elastomers is assumed unavoidable due to physical

and thermodynamic interactions. On decompression, the blistering or rupturing

is initiated at naturally occurring flaw sites. Resistance to blistering is a

function of pressure, flaw size, and elastomer shear modulus. Shear modulus is

the only variable we have control over. Figure 21, developed by Gent42 , is an

excellent story describing graphically the interactions of the three

variables. To vary shear modulus, the technologist can vary crosslink density

and filler type. Increasing molecular weight is predicted to increase fracture

energy43 and has been verified by functional and field testing.

13

REFERENCES

1T.W. Ray and C.E. Ivey, "Evaluation of Seal Materials for High Temperature H2SService", presented at Petroleum Society of CIM, June 13-16, 1978Calgary.

2T.L. Smith, "Strength of Elastomers-A Perspective", Polymer Eng. & Sci., 17, 129(1977); Rubber Chem. Technol., 51, 225 (1978).

3 L. E. Nielsen, "Mechanical Properties of Polymers", ASTM-D11.36 Geothermal Seals Task Group, Minneapolis, Minn., June 26, 1979 (with

revisions).

4"Engineering Properties Manual"-VITON � , E.I. Dupont, Fig. 5.3.3, Ch. 5, p.6.

5T.L. Smith, J. Polymer Sci., Part A13597 (1963); Rubber Chem. Technol. 37, 777(1964).

6T.L. Smith, J. Appl. Phys., 35, 27-36 (1964), Rubber Chem. Technol. 37, 792 (1964)

7R.F. Landel, R.F. Fedors, and J. Moacanin, "Recent Advances in Service LifePrediction", Appl. Polymer Symposium No. 22, 157-168 (1973) copyright 1973,John Wiley $ Son, New York.

8A.N. Gent, "Strength of Elastomers" in "Science and Technology of Rubber", F.R.Eirich, Ed.. Academic Press, New York, 1978, Ch. 10, p. 437.

9Reference 8, p. 434.

10Reference 8, p. 437.

11H.W. Greensmith, L. Mullins, and A.G. Thomas, "Rupture of Rubber", Trans. Soc.Rheology, 1960, 4, 179-89.

12A.Y. Coran, "Vulcanization" in "Science and Technology of Rubber", F.R. Eirich, Ed.,Academic Press, New York, 1978, Ch. 7, p. 292.

13 Reference 2, Figure 6.

14W.W. Schmiegel, "VITON Fluoroelastomer Crosslinking by Bisphenols", presented at theSouth German Meeting of the Deutsch Kautschuk and Gummi Gesellschaft, April28-29, 1977.

15M. Hudlicky, "Crosslinking of Perfluoroolefin Copolymers", presented at a meeting ofthe ASTM D11.36 Task Group on Geothermal Seals, June 27, 1978.

16K.L. Paciorek, "Chemical Crosslinking of Fluoroelastomers" in "Fluoropolymers", L.A.Wall, Ed., Wiley-Interscience, New York 1972, Ch. 10, p. 291-313.

14

17P.J. Flory, "Structure and Properties of Polymeric Materials", Stanford Univ.,

(January, 1977) NTIS AD-A042 643.

18M L. Dannis, "Rubbery Behavior-Simple Basics", Rubber Age, July p, 38 (1975).

19B. Vollmert, "Polymer Chemistry", Springer-Verlag, New York, 1973, p, 31.

20G. VerStrate, "Structure Characterization" in "Science and Technology of Rubber",

F. R. Eirich, Ed., Academic Press, New York (1978), Ch, 3, p. 112.

21PIBAL Report No. 69-38,"Thermo-Mechanics and Structure of Elastomers", Fig. III-1,

P111-5a, Brooklyn Poly Tech., Sept., 1969.

220. Kramer and J.D. Ferry, "Dynamic Mechanical Properties" in "Science and Technology of

Rubber", F.R. Eirich, Ed., Academic Press, New York 1978, Ch. 5, p. 193.

23 G. Schneberger, "Polymer Structure and Adhesives Behavior", Adhesives Age, April,

17 (1974) with modification.

24 Reference 19, p. 6.

25F.P. Baldwin, Rubber Chem. Technol. 42, 1171 (1969).

26 Reference 22, Ch. 5, p. 198.

27G.J. van Amerongen, Rubber Chem. Technol. 37, 1065-1152 (1964).

28R.C. Reid, J.M. Prausnitz, and T.K. Sherwood, "The Properties of Gases and

Liquids", 3rd Ed., McGraw-Hill, New York 1977, p. 545.

29J.H. Hildebrand and R.L. Scott, "Solubility of Non-electrolytes", Dover

Publications, New York 1950.

30J.H. Hildebrand, J.M. Prausnitz, and R.L. Scott, "Regular and Related

Solutions", Van Nostrand Reinhold Co., New York 1970.

31C.M. Hansen and A. Beerbower, "Solubility Parameters" in "Encyclopedia of

Chem. Tech.", 2nd Ed., John Wiley F, Sons, New York 1971.

32R.F. Fedors, "A Method for Estimating Both the Solubility Parameters and Molar

Volumes of Liquids", Polymer Science and Engineering, 14, 147 (1974) .

33A. Beerbower, private communication, Mar. 18, 28, 1980.

34J.H. Knox, "Molecular Thermodynamics", Rev. Ed., John Wiley $ Sons, New York 1971.

15

35D.W. Van Krevelin, "Properties of Polymers", 2nd Ed., Elsevier Scientific Pub.

Co., New York 1976, p. 439-442.

36 Reference 20, p. 85-86.

37Reference 8, p. 423.

38Reference 36, p. 125 (with modifications).

39Reference 8, Figure 7, p. 430.

40Reference 39, modified with Reference 28.

41Reference 8, p. 435.

42Reference 8, Figure 12, p. 432.

43Reference 8, p. 428.

16

NOMENCLATURE

C Celsius temperature

E Young's modulus

Eb elongation at break

G shear modulus

Gc critical fracture energy (critical strain energy)

K Kelvin temperature

Pc critical pressure (elastic instability)

ro radius-original (internal flaw or crack tip)

RR' plane of no net molal flow

Tg glass transition temperature

T temperature

V volume

δ solubility parameter

δd dispersion force parameter

δh hydrogen bond forces, parameter

δp orientation-electrostatic (dipole) parameter

ε strain, elongation %/100

εb strain at break

λ Extension ratio (length of the stretched specimen per unit

initial length)

λb Extension ratio at break

σ stress (load per unit initial area)

σb stress at break

ABBREVIATIONS

FKM hydrofluorocarbon elastomer (D1418)

FFKM perfluoroelastomer (D1418 prop.)

FKM1 low fluorine content

2 medium fluorine content

3 high fluorine content

FKM1A low viscosity (molecular weight)

B medium viscosity

C high viscosity 17

TABLES

Table I - Strength of Elastomers

Table II - Effect of Dispersion Forces, n-Alkanes

Table III - FKM Elastomers-Wt. % Fluorine

Table IV - Glass Transition (Tg) vs. Fluorine Wt.-%

Table V - FKM - Structural Repeat Units

Table VI - FKM Solubility Parameters

Table VII - Gas Solubility Parameters

Table VIII - Size Relationship - Elastomers, Fillers

Table IX - Failure Modes - Equations - Preconditions

18

FIGURES

Fig. 1 Shear Modulus-Tg-Molecular Weight Relationship

Fig. 2 FKM Modulus vs. Temperature

Fig. 3 Failure Envelope Schematic

Fig. 4 FKM Stress-Strain-Temperature Plot

Fig. 5 FKM Failure Envelope

Fig. 6 Amorphous Elastomer-Non-reinforced-Fracture Energy

Fig. 7 Amorphous Elastomer-Reinforced-Fracture Energy

Fig. 8 Crosslink Density vs. Physical Properties

Fig. 9 Curative Levels vs. True Stress at Break

Fig. 10 Meyer-Ferri Experiment

Fig. 11 Methanol Model

Fig. 12 Methanol Structures

Fig. 13 CH4-CF4 Boiling Points

Fig. 14 Polymeric Structures

Fig. 15 Physical Volume vs. Temperature

Fig. 16 Tensile Strength vs. Degree of Polymerization

Fig. 17 FKM Molecular Weight vs. Physical Properties

Fig. 18 Physical Entanglements of Elastomers

Fig. 19 Cavity Expansion - Triaxial Tension

Fig. 20 Diffusion Fluxes - Plane of No Net Molal Flow

Fig. 21 Critical Pressure-Cavity Size-Shear Modulus Relationships

19

TABLE I

Strength of Elastomers

Elastomer type Source of strength

Single-phase non- Viscoelastic processes

crystallizable Molecular network

Orientation of chains

Filled, Increased energy dissipation

non-crystallizable Deflection and bifurcation

of microcracks Cavitation

Crystallizable Formation and deformation

of crystalline domains

Block copolymers Plastic domains

___________________________________________________________________TABLE II

Building Molecular Weight

By Adding CH 2 Groups

Chemical Common Molecular Physical

Formula Name Weight State

CH4 Methane 16 Gas

C2 H6 Ethane 30 Gas

C3 H8 Propane 44 Gas

C4H10 Butane 58 Gas

C5H12 Pentane 74 Liquid

C17H36 Kerosene 240 Liquid

C18H38 Paraffin 254 Solid-soft

C50H102 Hard Waxes 702 Solid-brittle

C100H202 LMW Polyethylene 1402 Solid-tough

TABLE III

FKM Elastomers - Wt. % Fluorine

Wt %- F Wt %- H Molal %- H/F

FKM 1 66.0 1.86 53.5

FKM 2 68.5 1.40 38.9

FKM 3a 69.4 1.24 34.0

3b 70.0 1.14 31.0

FFKM 4 73.9 0.0 0.0

TABLE V

Degree of Polymerization

FKM - Structural Repeat Units*

FKM 1 FKM2 FKM3 FFKM

Mooney Value:

a Low 750 900

b Medium 1100 1250 950 400-800

c High 2200

* estimated

________________________________________________________________

TABLE VI

FKM Solubility Parameters1,2

1 2 a3

b 4

δd 15.4 14.3 14.7 13.5 12.1

δp 6.8 5.1 5.7 4.7 2.9

δh 9.2 7.2 8.2 8.8 3.5

δ 19.1 18.8 17.8 15.7 12.9

δ = √ δ2

d + δ2p + δ2

h

1. based on Group Contributions (unpublished)

A. Beerbower 3/18/80

2. CGS Conversion (÷) 2.046

22

TABLE VII

Gas Solubility Parameters

C02 H2S CH4 N2 He

δd 11.1 19.4 9.2 8.1 3.3

δp 7.2 8.2 0 0 0

δh 7.2 8.2 0 0 0

δ 15.1 22.8 9.2 5.1 3.3

1. A. Beerbower 3/29/80

2. STP - 25°c

3. CGS Conversion (÷) 2.048

23

TABLE IX

Failure Modes - Equations – Preconditions

Failure Mode Failure Equation Elastomer Preconditions

Blistering PC - 5E/6 Lower durometer,lightly crosslinked

elongation > 200% highly elastic

gas supersaturated

Lenticular (rupture) Pc = 4/3 σbλb High durometer, tightly crosslinked

elongation <100x, linearly elastic

1. A. Gent, Conversation 3/31 /80

24